Genetically heterogeneous (UM-HET3) mice produced as the progeny of (BALB x B6)F1 dams and (C3H x DBA/2)F1 sires will be used to map quantitative trait loci (QTL) that modulate longevity, age-sensitive traits, and early-life predictors of late-life events. Aim 1 will exploit an existing database, accumulated in the first 10 years of this project, which contains genetic data and phenotypic measures (of bones, hormones, immunity, weight, life span, and cause of death) for 1200 to 1800 mice, to evaluate new statistical methods for finding (a) epistatic interactions among QTL; (b) QTL that modulate relationships among traits, and (c) QTL that modulate variance in one or more traits. Aim 2 will use high resolution SNP genotyping to provide more accurate localization of QTL already localized by genome-wide scans. Aim 3 wilt produce and evaluate a new group of 600 UM-HET3 mice, first using a 300 - 400 SNP genome scan, and then, for selected traits, at higher resolution. Each mouse will be tested for a wide range of traits, including growth and maturation rates, T cell subsets, geometry and fragility of femur and vertebrae, multiple hormones, cataract severity, cause of death, and non-lethal pathological findings, as well as longevity. Most of these traits will be measured at two or more times in adult life, to allow searches for (a) QTL whose actions are restricted to specific times of the life span, and (b) QTL which influence the rate of change in age-sensitive outcomes. The data produced in Aim 3 will also be evaluated using the biostatistical methods developed in Aim 1, and QTL of special interest localized more precisely using the high resolution SNP methods of Aim 2. Aim 4 will ask which traits, measured in young or in middle-aged adults, best predict longevity and the timing of late life declines in immunity, endocrine pattern, bone status, and lens clarity. Early life predictors will provide physiological clues to the developmental events that influence aging in mammals, and late life predictors can be used as biomarkers to track the pace of aging in individual mice. This program should produce a detailed map of the genes that modulate individual aspects of aging in genetically heterogeneous mice, and allow a search for QTL that influence rate of change in multiple age-sensitive domains as well as life span and disease susceptibility.